AI and IOT in Renewable Energy

Table of Contents

1. Frontmatter

2. Chapter 1. A Day-Ahead Power Output Forecasting of Three PV Systems Using Regression, Machine Learning and Deep Learning Techniques

   Muhammad Naveed Akhter, Saad Mekhilef, Hazlie Mokhlis, Munir Azam Muhammad

   Abstract

   The forecasting of output solar power improves the quality, reliability and stability of power system. The aim of this research is day-ahead prediction of PV output power for 3 solar systems. The three PV systems are polycrystalline, monocrystalline and thin-film systems. A deep learning technique (RNN-LSTM) is proposed for day-ahead prediction of solar power output. The regression [GPR, GPR(PCA) and machine learning [SVR, SVR(PCA)] techniques are also developed. The forecasting accuracy is compared based on accuracy measurement parameters such as RMSE, MSE, correlation coefficient (R) and coefficient of determination (R²). One-year data for 2016 is considered for analysis. 70% of data is utilized for training and 30% for validation and testing. It is found that deep learning technique has better forecasting accuracy than other developed techniques in terms of lower (RMSE, MSE) and higher (R, R²), for day head forecasting of PV power output.

3. Chapter 2. Internet of Things and Internet of Drones in the Renewable Energy Infrastructure Towards Energy Optimization

   Ashok G. Matani

   Abstract

   A significant growth in implementing various renewable energy systems is observed throughout the world. Variable renewable electricity (VRE) sources such as solar PV and wind power have gained attractive investments in many countries, resulting in rapid growth in the installed capacity of these green sources of energy. The contribution of these variable renewable electricity energy sources had produced around 8.7% of global electricity as compared to 27.3% of all renewable energy sources at the world level. Therefore, there is an urgent need to improve the power system flexibility as the progress of the integration of variable renewable electricity energy sources. International efforts to meet renewable energy deployment and energy efficiency measures are resulting in a safe and reliable manner of renewable energy, thereby, resulting in minimized environmental, climate impacts, air quality improvement, good public health, and increased jobs and economic growth, increased grid reliability as well as lower energy costs on a household, corporate and national levels, The joint efforts by various institutions, corporations, governments, and non-governmental organizations (NGOs) has resulted in enhancing world level energy efficiency highlighting the potential to significantly minimization of greenhouse gas emissions on the earth. This paper highlights the latest developments in implementing Internet of Things (IoT) GPS and GIS tools and applications in energy sector in various parts of the world.

4. Chapter 3. Reinforcement Learning Algorithm to Reduce Energy Consumption in Electric Vehicles

   Manavi Shukla, Mandeep Singh Burdak

   Abstract

   This chapter consists of the analysis, design and testing of a reinforcement learning algorithm that is used to monitor the fuel efficiency in different environmental and terrain conditions to provide an optimized velocity, which if driven upon can improve the energy consumption behavior of an electric vehicle. To monitor the effects of the algorithm, a Simulink Electric Vehicle Model is used with parameters of a normal Sedan Size car with a lithium-ion battery energy source. First, the various factors that contribute to energy losses in an electric vehicle are defined. Then, the model-free Q-learning technique is explained. Finally, the tested energy consumption results are discussed, and further scope of this algorithm is mentioned.

5. Chapter 5. Simulation and Performance Analysis of Standalone Photovoltaic System with Boost Converter Under Irradiation and Temperature

   Kritika Khandelwal, Chetan Jain, Sunil Agarwal

   Abstract

   Renewable energy technology is the advanced technology capable of satisfying the problems of energy crisis that the world is facing as well as meeting the future energy demands due to the availability of energy source in infinite quantity in the atmosphere. This chapter presents a simulation and performance survey of the standalone photovoltaic (PV) system with boost converter under irradiation and temperature and in order to seize the utmost power at output Perturb and Observe (P&O) Maximum Power Point Tracking Algorithm (MPPT) is used. The output results are obtained and analyzed at different irradiations and temperature parameters. The proposed model is outlined and simulated in MATLAB/SIMULINK R2015a software.

6. Chapter 6. Analysis of Variation in Locational Marginal Pricing Under Influence of Stochastic Wind Generation

   Poonam B. Dhabai, Neeraj Tiwari

   Abstract

   This chapter presents an approach for analysis of variation in Locational Marginal Pricing (LMP) due to integration of stochastic wind generation in stabilized grid network by using Probabilistic Optimal Power Flow (P-OPF) in MATPOWER environment. The competitive market led to the assessment of the nodal price values for bidding the next MW charges. Consequently, the precise estimation and analysis of LMP values become a challenging task in the presence of dubious wind generation. LMP at any power transport bus ‘z’ constitutes of: base LMP, LMP due to losses at bus z and LMP due to congestion at the same bus. As LMP itself is composed of congestion cost factors at an individual bus, it not only aids in the determination of Network Rental (NR), yet in addition, has a significant part in the determination and management of congestion from an economical perspective within the system. Therefore presented work underscores a methodology to analyze the variation in LMP values highlighting the congestion scenario. The analysis is carried out statistically and power flow run. This analysis is performed on the wind data set obtained from Indian Meteorological Department (IMD) section, Pune, India, on a standard IEEE 30 bus test system.

7. Chapter 7. Optimal Integration of Plug-in Electric Vehicles Within a Distribution Network Using Genetic Algorithm

   Sakshi Gupta, D. Saxena, Mukesh Kumar Shah, Rajeev Kumar Chauhan

   Abstract

   This chapter mainly discusses and analyzes two smart charging approaches with objective functions as optimization of total daily cost (TDC) acquired by the charging facilitator and peak to average ratio (PAR), in that order to examine the effect on electric vehicle (EV) charging from both commercial and technical aspects. The proposed approaches are then executed on the industrial nodes of an 11 kV 37-bus distribution system to study the impacts as we increase the percentage penetration. Any system can accommodate a limited amount of EV penetration into it after which voltage and loading limits start to exceed. In this manner, here we are attempting to assess the most extreme conceivable PEV penetration by which the distribution system can entertain relating to both procedures.

8. Chapter 8. Frequency Control of 5 kW Self-excited Induction Generator Using Gravitational Search Algorithm and Genetic Algorithm

   Swati Paliwal, Sanjay Kumar Sinha, Yogesh Kumar Chauhan

   Abstract

   For harnessing the wind energy, self-excited induction generator is becoming more popular in today’s scenario. Nonlinear loads lead to major drawbacks in self-excited induction generator such as poor voltage, frequency regulation and reactive power consumption. This poor voltage and frequency of SEIG depends on many factors like types of load, capacitance involved for reactive power compensation and prime mover speed. The improved performance of SEIG can be obtained by using steady-state analysis of equivalent circuit and usage of optimization techniques in SEIG machine. The main objective of this chapter to select the values of shunt and series capacitances at specified speed in order to achieve an optimum frequency regulation using gravitational search algorithm (GSA) and genetic algorithm (GA). GSA works on Newton’s law of gravity, whereas GA follows the steps of selection, crossover and mutation. Both the techniques are based on heuristic approach of gbest and pbest. Therefore, this study is carried out on an objective function of relative mean error of frequency regulation. Correspondingly, the minimum fitness is calculated for resistive and resistive-inductive load. The improved performance validates both the optimization techniques.

9. Chapter 9. Cloud Based Real-Time Vibration and Temperature Monitoring System for Wind Turbine

   Subharshi Roy, Barnali Kundu, Debanjan Chatterjee

   Abstract

   In the present scenario, Renewable Energy requires real-time condition monitoring for their uninterrupted performance. Wind turbines are often subjected to huge mechanical, and thermal stresses which in turn result in causing faults. In this paper, a Cloud-based Real-time Monitoring System (CRMS) has been developed for the early detection of a problem and identify the need for maintenance before a wind turbine fails. CRMS associated with vibration sensor and temperature sensor can easily detect the fault and alarming system indicates the operator personnel about the abnormal state of the motors in the industrial plant. With real-time data monitoring system and LabVIEW, it enables the detailed spectral analysis of the system. A wireless sensor networks are included in this research work for a real-time condition monitoring. Therefore, the authors of this paper have developed a prototype which can provide smart maintenance to elongate the life of wind turbine and prevent the harm of nearby equipments.

10. Chapter 10. Smart Solar-Powered Smart Agricultural Monitoring System Using Internet of Things Devices

    Paras Patel, Anand Kishor, Gitanjali Mehta

    Abstract

    India is the fastest-growing big financial system in the world, with a massive population, useful demographics, and excessive catch-up potential. In India, agriculture is a primary activity, around two-third of India's population remains dependent upon agriculture. In developing nations, farmers aren't using smart agricultural technique but if they begin the use of smart agricultural technique with the assist of this technique, they can produce good yield crops, wide range of development on the agriculture, and can make superior amount of profit. To reduce long-time expenditure in agriculture, use of renewable energy is important for that smart solar is the primary energy which may be used.